Remove unnecessary crazy_linker order_profiling logic

Remove unnecessary crazy_linker order_profiling logic

The logic is unnecessary because the crazy linker is not used for chrome_apk
when order_profiling=1.

BUG=531278
TEST=None

Committed: https://crrev.com/9b8f934f3e079cd9412337df9fbab5a8ea6cfd9e
Cr-Commit-Position: refs/heads/master@{#350839}

[pkotwicz, 2015-09-23 02:29:43]

pkotwicz@chromium.org changed reviewers:
+ pasko@chromium.org

[pkotwicz, 2015-09-23 02:29:43]

pasko@ PTAL

[pasko, 2015-09-23 09:42:47]

pasko@chromium.org changed reviewers:
+ torne@chromium.org

[pasko, 2015-09-23 09:54:20]

Based on history of crazylinker, we might want to get back to using our forked
crazylinker if we find a new crazy problem in the android linker, so changing
this code now may cause extra conflicts when cherry-picking later.

Adding torne@ to reviewers, as he knows better about all the current
configurations of the crazy linker. If the change is good for him, and passes
the test that I provided below, then I'm glad to remove this.

> The crazy linker is not used for chrome_apk when order_profiling=1 because
> use_chromium_linker=0 for chrome_apk whenever the static_library build is
> used.

The crazy linker and order_profiling are android only, and we never use
static_library build to run Chrome on Android, so the above does not provide me
with any useful information.

> order_profiling=1 requires the static_library build.

This is not true. I am not sure why you came into this conclusion, maybe you
want to look at downstream GYP files around their use of order_profiling?

https://codereview.chromium.org/1366433004/diff/1/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/1366433004/diff/1/build/common.gypi#newcode4594
build/common.gypi:4594: # don't want to instrument it.
is this comment no longer true?

The proper test of this functionality would be to:
* build downstream Chrome for Android order_profiling=1
* throw an intent at it to load a google.com
* wait for 1 minute

If it does not crash, we are good to go

[pkotwicz, 2015-09-23 18:00:54]

pasko@ can you please take a look?

I meant component=static_library. orderfile_generator.py builds with
component=static_library (which is the default)

https://codereview.chromium.org/1366433004/diff/1/build/common.gypi
File build/common.gypi (right):

https://codereview.chromium.org/1366433004/diff/1/build/common.gypi#newcode4594
build/common.gypi:4594: # don't want to instrument it.
I removed the comment. Thank you for noticing! We never use the crazy_linker
when building chrome_apk with order_profiling=1 so we do not need to modify the
crazy_linker target at all.

I have checked that downstream Chrome for Android does not crash with these
changes when built with order_profiling=1

[pkotwicz, 2015-09-24 12:51:06]

pasko@ Ping!

[pasko, 2015-09-25 12:20:48]

On 2015/09/23 18:00:54, pkotwicz wrote:
> pasko@ can you please take a look?
> 
>
> use_chromium_linker=0 for chrome_apk whenever the static_library build is
> used.
[snip]
> I meant component=static_library. orderfile_generator.py builds with
> component=static_library (which is the default)

So you are saying that use_chromium_linker=0 whenever we build Chrome.apk.
I am having a hard time agreeing with this. What do you think about setting
use_chromium_linker to 1 in chrome_android.gypi?

[pasko, 2015-09-25 13:35:38]

lgtm because chrome_apk_use_chromium_linker is set to 0 in chrome_apk.gypi when
order_profiling!=0

[pkotwicz, 2015-09-25 13:39:13]

pkotwicz@chromium.org changed reviewers:
+ agrieve@chromium.org, yfriedman@google.com

[pkotwicz, 2015-09-25 13:39:14]

agrieve@ for build/
yfriedman@ for base/

[Yaron, 2015-09-25 13:50:16]

yfriedman@chromium.org changed reviewers:
+ yfriedman@chromium.org

[Yaron, 2015-09-25 13:50:16]

base/ lgtm

[agrieve, 2015-09-25 14:14:32]

On 2015/09/25 13:39:14, pkotwicz wrote:
> agrieve@ for build/
> yfriedman@ for base/

lgtm

[pkotwicz, 2015-09-25 14:20:00]

The CQ bit was checked by pkotwicz@chromium.org

[commit-bot: I haz the power, 2015-09-25 14:20:18]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1366433004/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1366433004/20001

[commit-bot: I haz the power, 2015-09-25 15:57:20]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-09-25 15:58:37]

Patchset 2 (id:??) landed as
https://crrev.com/9b8f934f3e079cd9412337df9fbab5a8ea6cfd9e
Cr-Commit-Position: refs/heads/master@{#350839}